M13 – Great Cluster in Hercules – And an experiment in LRGB
I have, this week, been experimenting with LRGB photography, using a Mono CCD camera and coloured filters. This will explain the lack of a writing post this week and, indeed, the lack of any writing. However, like hen’s teeth, clear skies are a rare occurrence, and I have been taking advantage of them this week.
As I am new to this form of imaging I thought I would try for a relatively “easy” target – i.e. one that was in a good position and relatively bright. The spring skies are getting darker later and later, and the time available for imaging is limited, so I didn’t want to go for something too faint.
M13, the Great Globular Cluster in Hercules is one of the most popular clusters in the night sky and a good target for testing out a new camera. I really just wanted to make sure I could get everything to work and then process the images into one LRGB colour image, so I wasn’t planning a long session. I decided to go for 1 hour of Luminance and 15 minutes each of Red, Blue and Green Channels.
The Luminance filter lets all wavelengths of light through and provides the majority of data and detail for the main image – essentially it produces a normal mono image. The Red, Blue and Green channels can be “binned” which means several pixels on the sensor are combined into 1 “superpixel” covering a larger area of the target. This allows the data to be captured much faster in the colour channels. It does mean the quality of the image is lessened as the resolution is reduced but the main Luminance image collects the detail, so this is not a major problem, unless your pixels are very big in the first place. Collecting data at 2×2 binning combines the data from 4 pixels (3×3 9, etc etc,) and reduces the time required by a factor of 4, so for 60 minutes of Luminance, we only need 15 minutes each of Red, Blue and Green – so an hour and 45 minutes in total. Higher binning levels will reduce the time further, but also greatly reduce the data quality and is probably not suitable for my camera.
Once set up and aligned, I found M13, got focussed and guiding and then set the camera to take 20 x 180 second images, followed by several BIAS and FLAT images for later calibration. I did not take the usual DARKS on this run as the new camera is actively cooled, meaning I can control the temperature down to 30 degrees below ambient temperature. DARKS are designed to help calibrate your images and account for hot pixels and other problems caused by a heated sensor. With the on-camera cooling, DARKS are, theroretically, not required. I may well experiment with them in the future but, on this occasions, I decided not to. However, with the cooling available, the DARKS can now be taken at any time before or after the imaging session as the cooling allows me to easily replicate the temperature of the imaging run – something not so easy to do with a non-cooled DSLR!
After the Luminance issues were complete, I did the same for the Red, Green and Blue channels, selecting the appropriate filter as I went along. I have a manual filter wheel that holds 5 filters. I currently only have 4 slots filled with the LRGB filters, but it allows me to easily select the filter I want before each run.
At least that was the plan.
Somehow I managed to forget to change the filter wheel from Green to Blue and ended up with two sets of Green images, just as M13 disappeared behind the house! Very annoying… so I had to wait a couple of nights for another clear sky so I could get the final 15 minutes of Blue images, along with the relevant calibration frames. Unfortunately the night wasn’t as clear as it was supposed to be and many of the images were “tainted” by cloud. Although still visible, the clouds diffused the Blue channel light which affects the final picture as you will see below.
Anyway, I then had 4 sets of data for my LRGB channels, along with the required calibration frames. Each set was combined, stacked and processed in as similar way as possible to ensure they were of similar quality using PIXINSIGHT. The process was fairly basic with just a CROP to get rid of the dodgy corners left over from stacking. DBE to get rid of any gradients/vignetting, a quick HISTOGRAM stretch to bring the data out and then HDR Multiscale Transformation to bring out some detail in the core of the cluster. The colour channels also had to be resized to match the Luminance channel as the binning process also reduces the size of the image.
I processed the Luminance channels (mono) first off, before I had the Blue images, just to see what the data was like and, for just an hour, I was suitably impressed. I never got anywhere near this with a DSLR.
Next step was to combine the RGB channels using CHANNEL COMBINATION in Pixinisight and I then applied this to the Luminance Channel using the LRGB Combination Process. Finally I took the image from Pixinsight and tweaked the levels and curves in Photoshop.
As I mentioned above, the clouds affected the quality of the blue channel, meaning the light was diffused around the brighter stars and you can see this in the final image below. However, I was still happy with the result as I was really just trying to see if I could “do it” and get four sets of images from a Mono camera and combine them into a colour image.
There is definitely colour in the image…. maybe not all the right colours in the right places with the right weights, but there is colour! As mentioned above, you can clearly see the effect of the clouds on the two stars with blue fringing which shouldn’t be there. Next time out I just need to make sure there aren’t any clouds about!
Which should be easy!
Larger images are in the Gallery!